Why low rail adhesion still presents a high risk
The Salisbury rail crash resulted from slipping wheels, but why is this still a problem on today’s railway, and what can be done about it? GARETH DENNIS digs into the detail.
A version of this article also appeared in Issue 946 (15 December 2021) of RAIL magazine.
Put aside the £286M a year costs associated with reduced performance. Pass over the £49M a year spent on prevention, management or cleaning. Indeed, ignore the £10M spent on component replacement. The images from the Salisbury rail crash should be enough to convince any of us that low adhesion and the wheel slip that it can cause is a problem the railway must take seriously.
To understand what we can do about it, it is first worth thinking about the science of friction, and how it applies to wheel-rail adhesion. Permit me to drag you back to A Level mechanics for a minute.
Imaging a block sat on a table. It has a weight, which is equal to its mass times the acceleration due to gravity. That weight is the force compressing the block’s lower surface against the surface of the table. If we push against the block, we will need to apply an amount of force that is equal to some percentage of its weight to get it to move. The “some percentage” bit is the coefficient of friction: as the coefficient of friction increases, the closer the applied force gets to the weight of the block. A low coefficient of friction means that very little applied force is required to get the block moving.
How is this relevant to railway wheels? Well, in the ideal condition, the contact between the wheel and the rail acts as a stationary block on a table: the wheel should not be rotating faster or slower than the passage of the surface of the rail underneath it. If it does, then the wheel will be slipping.
When accelerating or braking, a force is applied to rotate (or slow the rotation of) the wheels and so long as this force remains less than the axle weight times the coefficient of friction, there will be no wheel slip. Our block won’t slide across the table.
A full service brake application of a modern train can apply about 9% of a train’s axle weight (referred to as 9%g). The coefficient of friction between a dry steel train wheel and a dry steel rail usually sits between 30‑40% of axle weight, so braking presents no problem. Even in severely wet conditions where adhesion (the percentage coefficient of friction versus axle weight) can drop to 10‑20%, braking is generally unproblematic though accelerating trains from a standstill can cause issues.
Wheel slip under braking occurs where adhesion drops below the braking force being applied — our block starts sliding across the table when the force we are applying exceeds the coefficient of friction times the weight of the block. If the driver requests a full service brake application of 9%g and adhesion at the rail head is 9%, then the wheel will start to slip.
Leaf mulch mixed with rust particles and compressed to a paint-thin paste can reduce adhesion to as little as 1%. In other words, leaves on the line can reduce the coefficient of friction between the wheel and the rail by more than 95%. You can see the problem.
There is no end of ideas to tackle leafy low adhesion. A review by the Adhesion Working Group a few years ago listed over fifty possible approaches at varying levels of viability (and wackiness!), and they can be split into three categories: reducing risk from low adhesion events, improving adhesion at the rail head and avoiding leaves reaching the rail in the first place.
Possibly best known is the fleet of rail head treatment trains of various shapes and sizes. They use a pressurised jet of water to blast away the leaf mulch then apply Sandite — a mixture of fine-powdered clay, sand, antifreeze and steel particles first deployed by British Rail in the 1970s — to improve adhesion and signalling performance. In trouble spots, maintenance teams scrub rail heads and apply Sandite manually.
It is also common these days for trains to have wheel slide protection and driver advisory systems that provide a warning when adhesion is low to reduce the reliance on defensive driving techniques. As train control software and hardware gets more complex, these subsystems become more advanced too.
Behind the scenes, Autumn timetables are planned to extend journey times to allow drivers to brake and accelerate more gently and avoid wheel slip. Less obvious still are the forecasters in both the Met Office and Network Rail predicting when and where windy weather will accelerate leaf fall requiring additional mitigation.
There are also extensive research efforts investigating new or adapted approaches to removing leaves or leaf mulch and the use of microwave-superheated steam or cryogenic dry ice has found some success. Magnetic track brakes have been identified by the RSSB as a viable means of improving braking performance on slippery rails as well.
Collaboration between the Met Office, universities and the RSSB has resulted in improvements to forecast relevance and resolution too, enabling Network Rail to get ahead of leaf fall and informing drivers in real-time when their train might encounter low levels of rail head adhesion.
Fundamentally though, innovations in forecasting, cleaning, train control and braking are no match for keeping leaves (and indeed trees) off the line in the first place — and for the most part that means firing up the chainsaws.
Government must fund maintenance accordingly, as clearance work is complex and labour intensive, but they also need to support Network Rail to tackle the inevitable local political obstacles to felling. The myths about trees being vital to earthwork stability need to be busted, too (see my pieces in RAIL912 and RAIL946).
As we squeeze ever-more out of our existing lines, particularly considering the government’s recent cancellation of new infrastructure, we will require trains to be travelling in closer proximity and with sharper acceleration and deceleration. This, alongside the increased prevalence of extreme weather events associated with climate change, will only heighten the risks associated with leafy low adhesion.
If we want Britain’s railways to be resilient, reliable and safe, then (the benefits of other methods notwithstanding) we will need to pursue a far more aggressive approach to lineside vegetation management.
